implement-reviewed-issue
Skill2 starsChanged 27 days agoMITView on GitHub
Implement an issue, run a sub-agent review after development, resolve findings, then commit, push and create or reuse its pull request. Use when the user requests this complete reviewed issue-to-PR workflow.
In this repository4 other files it gives its agents.
The file
Files
Implement, Review and Publish an Issue
Accept an issue number or URL. Complete the issue using implement-issue, including its same-checkout semantic branch, documentation, drift reconciliation and verification requirements.
After implementation and relevant checks are complete, delegate an independent code review to a sub-agent. Supply the issue acceptance criteria, repository and branch/base, the full intended diff (including new files), applicable instructions and validation results. Ask for actionable correctness, regression, coverage and documentation findings with file/line references. The reviewer should inspect the change without editing it or publishing anything. If sub-agents are unavailable, report that the requested independent review is blocked; do not present a self-review as its replacement or publish before that requirement is resolved.
Evaluate the review findings against the issue and repository contracts. Fix valid findings, run checks covering the corrections, and request a focused follow-up review when substantive changes warrant it. Resolve any material disagreement explicitly. Do not publish with an unresolved correctness blocker.
Once development, verification and review are finished, use commit-push-pr to commit the intended changes, push the issue branch, and create or reuse its PR. Invoking this complete workflow authorizes these publishing steps; respect any narrower user instruction. Follow that skill's lookup across all PR states, destination checks and attachment behavior. Do not merge, deploy or post separate issue comments without separate authorization.
Report the issue, branch, implemented behavior, checks, review outcome, commit and PR link, plus any remaining limitations. A blocked review or publishing step must be stated accurately while preserving completed local work.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.